Acoustic Seafloor Classification Using the Weyl Transform of Multibeam Echosounder Backscatter Mosaic

Abstract

The use of multibeam echosounder systems (MBES) for detailed seafloor mapping is increasing at a fast pace. Due to their design, enabling continuous high-density measurements and the coregistration of seafloor’s depth and reflectivity, MBES has become a fundamental instrument in the advancing field of acoustic seafloor classification (ASC). With these data becoming available, recent seafloor mapping research focuses on the interpretation of the hydroacoustic data and automated predictive modeling of seafloor composition. While a methodological consensus on which seafloor sediment classification algorithm and routine does not exist in the scientific community, it is expected that progress will occur through the refinement of each stage of the ASC pipeline: ranging from the data acquisition to the modeling phase. This research focuses on the stage of the feature extraction; the stage wherein the spatial variables used for the classification are, in this case, derived from the MBES backscatter data. This contribution explored the sediment classification potential of a textural feature based on the recently introduced Weyl transform of 300 kHz MBES backscatter imagery acquired over a nearshore study site in Belgian Waters. The goodness of the Weyl transform textural feature for seafloor sediment classification was assessed in terms of cluster separation of Folk’s sedimentological categories (4-class scheme). Class separation potential was quantified at multiple spatial scales by cluster silhouette coefficients. Weyl features derived from MBES backscatter data were found to exhibit superior thematic class separation compared to other well-established textural features, namely: (1) First-order Statistics, (2) Gray Level Co-occurrence Matrices (GLCM), (3) Wavelet Transform and (4) Local Binary Pattern (LBP). Finally, by employing a Random Forest (RF) categorical classifier, the value of the proposed textural feature for seafloor sediment mapping was confirmed in terms of global and by-class classification accuracies, highest for models based on the backscatter Weyl features. Further tests on different backscatter datasets and sediment classification schemes are required to further elucidate the use of the Weyl transform of MBES backscatter imagery in the context of seafloor mapping.

The Suitability of Machine-Learning Algorithms for the Automatic Acoustic Seafloor Classification of Hard Substrate Habitats in the German Bight

“…Instead of artificially generating image patches, the authors in Ref. [2] artificially generated new grab sample locations. They assume that inside a manually specified area around the original location of a given grab sample, the seafloor is made of the same components as the actual sampling locations (see Figure 4,left).…”

“…Prediction maps of the seafloor surface characteristics in the study sites were computed with three different supervised machine-learning algorithms: Support vector machines with a linear kernel (SVM-L) (e.g., [33,34]), random forests (RF) (e.g., [2,4,5,9,10,[35][36][37]), and convolutional neural networks (CNN) (e.g., [1,[38][39][40]). SVMs have been widely used in remote sensing (see [41] for a detailed review) and seafloor classification tasks.…”

“…In order to meet the challenges of an increasing amount of data and the necessary cycles of habitat evaluations in the future, greater automation and standardization of the classification process are required. Recent years have seen progress in automated seafloor classification, with various machine-learning classification methods employed to enhance the identification of seafloor characteristics using hydroacoustic data, oceanographic variables, and ground-truth samples (e.g., [1][2][3][4][5][6][7]). Although comparative studies of some of the applied algorithms exist (e.g., [8][9][10]), these studies either lack visual representation of model performances or do not include sophisticated deep learning techniques.…”

Seagrass Habitat Suitability Models using Multibeam Echosounder Data and Multiple Machine Learning Techniques

“…The predictive value of these variables is better than that of MBES data (i.e., bathymetric map and backscatter mosaic) as they provide the detailed information on seabed topography and substrata [51,52]. Among the various predictors, the majority of previous research typically used slope, curvature, eastness, and northness derived from bathymetric map and Gray Level Co-occurrence Matrix (GLCM) texture features [53][54][55][56][57] and Angular Range Analysis (ARA) parameters derived from backscatter mosaic [58][59][60]. Though bathymetric map has higher importance in modelling seagrass habitats [12,[61][62][63][64], other backscatter predictors may also contribute significantly to improve the model [62,65].…”

A Survey of Seafloor Characterization and Mapping Techniques

“…Zhao et al [81] used the Weyl Transform to characterize backscatter imagery provided by MBES obtained in the Belgian section of the North Sea. The authors' proposal focused on the textural analysis of seafloor sediments and used the Weyl Transform to obtain the features.…”